對信用卡公司而言,如何能同時兼顧提升信用卡刷卡總金額與控制呆帳成本,總是一個兩難的課題。 信用卡刷卡總金額的提升是信用卡公司最重要的經營目標,但是信用卡公司也不樂見持卡人在大筆支出後,因為無力一次清償,滾入循環後卻愈拖愈久,最後無力償還變成呆帳。因此信用卡公司提供優惠利率,為持卡人規劃償債的期限,讓持卡人在一定的期數內順利還完,信用卡公司也可降低未來的呆帳風險。 本篇研究目的即是利用羅吉斯迴歸探討影響信用卡消費者申請信用卡帳款分期付款因子,並據以建立羅吉斯迴歸模型,計算客戶的申辦機率,以辨識可能申辦信用卡帳款分期付款的目標客戶,提高信用卡公司行銷的效率,降低行銷成本,提高產品利潤率與市場競爭力。 研究發現消費者在申辦信用卡帳款分期付款之前,會考量個人可取得的授信最低利率成本,包含信用貸款利率與信用卡循環利率成本等;同時,個人的特質與消費、繳款習慣、往來情形等也都是影響申辦的因子之一,例如:性別、年齡、持卡年資、預借現金使用情形、近期消費金額與利息變動情形等,也都是影響申辦信用卡帳款分期付款的因素之一。 使用羅吉斯迴歸分析方法,除可以讓使用者以較低的成本尋找影響反應變數的自變數資訊外,另因羅吉斯迴歸分析所得的預測值是機率值,業者可視自身現行可用資源,自訂臨界值,調整、篩選欲納入行銷的目標客戶名單,藉以改善行銷名單的品質,提高行銷成功率以達到下降行銷成本的目的。
For credit card companies, how to enhance generating revenue by encouraging card holders to use credit cards and, meanwhile, to control the costs of bad debts is always a tough issue. Although raising the total revenue of credit card is companies’ most important business goal, credit card companies may not want cardholders to lose ability to discharge their debts after a large consumption. To solve the problem, credit card companies offer preferential interest rates and repayment period for cardholders to successfully resolve their debts. Consequently, credit card companies may reduce the risk of bad debts expense in the future. The purpose of this study is to find out the factors which have impact on consumers to apply for credit card bill installment. The methodology used in this study is logistic regression. In addition, this study uses those factors found to establish logistic regression models to calculate the probability in order to help credit card companies target potential customers who would apply for credit card bill installment. This information may help credit card companies to improve marketing efficiency, lower marketing costs, and improve product profitability and market competitiveness. This study found that customers will consider the lowest interest rates of all kinds of their loans, such as interest rate of installment and interest rate of credit card bill before they apply for credit card bill installment. This study also found that personal characteristics, such as gender and age, and consumption habits, bank records, the duration of holding a credit card, cash advance usage, and interests of credit card will also have impact on cardholders’ tendency to apply for credit card bill installment. Logistic regression analysis can identify the factors which have impact on the application rate of credit card bill installment, and it also lowers costs of study for users. Furthermore, the response variables of logistic regression is a probability, so researchers can decide which cutoff value will be more suitable and will help to adjust their selection of target customers list. In other words, the advantages of this study are to help companies to select potential customers for marketing in order to improve the success rate of marketing and decrease marketing costs.